Dynamic cancer drivers

Dynamic Cancer Drivers identify genes that causally drive one or more bio-pathological transitions during cancer progression by integrating pseudotime-informed causal inference applied to single-cell and bulk sequencing data.

Key Features:

• Causal Inference Framework: Employs causal inference to assess genes' effects on biological transitions during cancer progression.
• Pseudotime Analysis: Uses pseudotime trajectories to infer latent progression and locate critical moments of transition.
• Critical Event Detection and Causal Impact Assessment: Detects critical events where deviations from normal processes occur and assesses genes' causal impact on transitions after those events.
• Data Modalities: Applies to single-cell and bulk sequencing datasets, including applications in breast cancer.
• Improved Performance Over Static Methods: Demonstrates superior driver discovery compared with traditional static approaches by integrating temporal dynamics.

Scientific Applications:

• Enhanced Cancer Driver Discovery: Enables discovery of driver genes associated with specific bio-pathological transitions that static analyses may miss.
• Understanding Cancer Progression: Reveals sequential and causal relationships between genetic alterations and process-level transitions in cancer progression.
• Personalized Medicine: Identifies dynamic drivers specific to patients or subtypes to inform personalized treatment strategies.

Methodology:

Inferring pseudotime trajectories to model progression; identifying critical events where significant deviations from normal processes occur; and assessing the causal impact of genes on those transitions post-critical event.